Integrating artificial intelligence with non-destructive experimental methods for effective food analysis: a case...

Spices are essential in daily life, serving both culinary and medicinal purposes and have profoundly influenced human history, culture, economy, and health. However, a key challenge associated with these natural ingredients is the limited information available for their identification and quality assessment. The critical issues of spice processing and authentication, emphasize the need for a robust digital authentication system. Hence, there is a need for an innovative approach leveraging artificial intelligence and machine learning to enhance spice authentication. The integration of food computing and non-invasive experimental methods has emerged as a significant approach to enhance the analysis and quality control of spices. This review explores how advanced computational techniques, such as artificial intelligence, can be combined with non-destructive experimental methods such as Near-infrared spectroscopy, Fourier transform infrared spectroscopy and Hyper Spectral Imaging to achieve more comprehensive, efficient, and accurate spice components analysis. Spices, with complex compositions and high economic, nutritional, functional, and sensory value characteristics, serve as an ideal case study using artificial intelligence, machine learning, and deep learning for the classification, authentication, and quality assessment of spices. It highlights the ability of food computing to process and interpret large datasets obtained from spectroscopy, chromatography, and sensory evaluation. With recent advances and case studies, this review emphasizes the potential of AI-integrated non-destructive experimental approaches to address current challenges in spice quality analysis. This ensures the authenticity, safety, and quality of spices supporting industry efforts to enhance product quality, reduce waste, and meet consumer demands.